DNA ComputingUladzislau Maiseichykau

Michael Newell

Overview● Introduction to DNA computing● History● Differences between DNA and traditional computing● Potential● Limitations● Constructing a DNA computing experiment● Early DNA computing● Modern DNA computing● Future of DNA computing

What is DNA Computing?DNA computing uses DNA, chemistry, and molecular hardware instead of traditional silicon-based circuits to run calculations and computations

Interest in DNA computing stems from the possibility of processing larger sets of data in a smaller space than silicon computers, and the fast-approaching physical limits to Moore’s Law

History● Initially developed by Leonard Adleman from

University of Southern California in 1994● Used DNA to solve the Traveling Salesman

problem● In 1997, researchers from University of

Rochester used DNA to create logic gates● Researchers from Weizmann Institute of Science

unveiled a programmable computing machine that used enzymes and DNA in 2002

● 2013, the first biological transistor, the transcriptor, was created

Comparison to Standard Computing

“DNA computers are unlikely to become stand-alone competitors for electronic computers.”

Smallest most efficient computers, modifies cells

Build items vs. raw mathematics

Potential● Ultra small devices● Self assembly (nano) - ex: Sierpinski triangle● Exotic and new molecular structures● Nano-Gold + DNA = 10 smaller details on chips

Limitations● Human involvement to set up each experiment/calculation● Response times can be as slow as minutes, hours, or even days● Results are much harder to read, often takes longer to find the correct

answer than to compute possible solutions

Constructing a DNA Computing Experiment

Travelling Salesman Problem

● Salesman needs to visit all cities taking the shortest route possible● First, fragments of DNA are created representing each city● Each of these fragments are able to bond with each other representing travel to a different city● The strands are mixed in a test tube and bonds are formed in seconds● A chemical filter process (lasting days) eliminates invalid or overly long solutions● The remaining strings of DNA are possible paths the salesman can take

We do not yet have the technology to read the strings of DNA efficiently, so this experiment remains a proof of concept

Early DNA Computing

● Too much lab work● Large number of answers, manually pick the right answers from them● 100 trillion answers in a second, most repeats or incorrect● Several dozen more lab procedures to sift through the answers

Modern DNA Computing

● DNA Tiles (Wang Tiles)● Edges (numbers and shapes)● Erik Winfree● Possible to convert to binary and sift through answers quickly● Not yet complete● If implemented properly, “several trillion multi-tile structures each of

which has carried out an orderly addition of three binary bits”

Future of DNA Computing● Transcriptors may make organic computers

possible in the future ○ Slower than traditional transistors, but can

work inside a body to monitor and affect surrounding cells

● Self-assembling nanomachines○ Researchers used enzyme encoding to

create Sierpinski triangles on DNA strands● DNA can be used to store enormous amounts of

data in a very small space

References● https://en.wikipedia.org/wiki/DNA_computing

● http://computer.howstuffworks.com/dna-computer.htm

● https://www.britannica.com/technology/DNA-computing

● https://www.usc.edu/dept/molecular-science/papers/fp-sciam98.pdf

● http://www.dna.caltech.edu/Papers/FOE_2003_final.pdf

● https://www.technologyreview.com/s/400727/dna-computing/● http://www.math.ubc.ca/~cass/courses/m308-02b/projects/schweber/tilings.html